Bulletin 142 - 2005 Aug 05

1. HPCCC SX-6 SUPER-UX upgrade to R15.1
2. New HPCCC Satisfaction Feedback Form
3. Use of do_tx7/do_sx6 and rsh commands - the -n option
4. Totalview class scheduled
5. TX7 software upgrade, and a new version of rsync on the TX7s
6. cherax, CSIRO Data Store, and www.hpsc.csiro.au downtimes
7. www.hpsc.csiro.au - userguide, and access to cherax man pages
8. HPCCC Seminar on SX-6 Issues

Information about the upgrade to SUPER-UX R15.1 is being made available at: http://www.hpccc.gov.au/hpccc/user_news_advice/ and http://www.hpccc.gov.au/hpccc/userdocs/index_user.shtml .

In preparation for the upgrade the HPCCC has recently completed

• changes to NQSII, ERSII, and resource groups
• installing updated cross-environments on the front-ends

On 9 August we will

• update c++ on the SX-6s and front-ends
• update nco, udunits and netCDF libraries

Installation of the updated versions of the MPI libraries, commands and daemons is delayed while we work on better ways to make different versions available.

At http://www.hpccc.gov.au/hpccc/userdocs/SuperUX_Node_Upgrade_to_15-1.shtml there is information about the timetable for nodes to be moved from R13.1 to R15.1.

From Wed 10 August until the upgrade is complete, users can submit jobs to nodes running R13.1 by using the usual queues, and to nodes running R15.1 by the usual queue names with a '3' suffix: e.g. bm3.

NEC documentation for R15.1 will appear at
http://www.hpccc.gov.au/hpccc/userdocs/index_user.shtml .

The Feedback Form available at
http://www.hpccc.gov.au/hpccc/helpdesk/feedback.shtml
has been upgraded with subcategories and topical comment boxes.

The HPCCC offers its thanks to everyone who has already responded, and encourages everyone else to make their views known.

HPCbull items

• 115.6. Use of the do_tx7 script
• 128.4. HPCCC SX-6 do_tx7 and TX7 do_sx6 utilities

We have had another recent case of jobs hanging when using these commands.

One way to reduce the chances of these hangs is to use an enhancement to these commands, by including the -n flag as the first option to the do_tx7 command, e.g.

 do_tx7 -n rcp myfile remote_machine:

The -n command should also be used in nearly all any scripted uses of rsh - see the rsh man pages for descriptions of the problems.

TotalView is the debugger of choice for parallel programming, and especially for MPI. It is available on the SX-6.

Joerg Henrichs (NECA) will conduct an SX-6 Totalview debugging seminar on Tuesday 11 October.

• 10:00-11:30 TotalView Lecture in the BMRC 9E Lecture Room
• 11:30-13:00 Hands On w/Joerg in the BMTC Computer Laboratory Room (11 terminals w/Xwin available)

Please sign up early so we can ensure our facilities meet your requirements.

To register contact Len.Makin@csiro.au, or 03 9669 8109.

The HPCCC expects to upgrade the operating systems on the TX7s in the fourth quarter of 2005. The upgrade will provide greater redundancy in access to disc units.

In response to a user request, a new version of the rsync utility has been installed on the TX7s. (The default version has not been changed.) The new version provides some extra facilities, such as working from a list of files.

To use rsync as a client you can set up your environment with:

    pkgenv rsync-2.6.6

    pkgenv rsync

To use rsync as a server, use the option

--rsync-path=/common/ia64/rsync-2.6.6/bin/rsync

Note: 'make test' indicated that there were some problems in handling of hard links. If you need to rsync hard links you should test carefully.

cherax, the CSIRO Data Store, and www.hpsc.csiro.au will be down on Saturday 13th August, and maybe part of Sunday 14th August.

New disc will be incorporated into the /cs/datastore and /work file systems, and configured to provide higher performance. The /cs/datastore will be dumped and re-loaded - this will take several hours.

Farrer will also be down, along with access to the CSIRO Data Store from burnet and nelson.

Further scanning of the documentation has revealed no user level issues as yet.

There is an upgrade to the Apache WWW server - WWW services may not be restored until Monday 15th August.

There are issues with SAMBA support: the experimental export of the /cs/datastore file system to the CSIRO Windows domain may not be restored until Monday 15th August. (Because of current unresolved problems, this experimental service is currently not available from \\cherax.hpsc.csiro.au\user but from \\farrer.hpsc.csiro.au\user.)

Information about the upgrades can be gleaned from our WWW pages at
http://intra.hpsc.csiro.au/user/userdocs/ax/

From the HPSC's WWW pages for users at http://intra.hpsc.csiro.au/, the Master Userguide at http://intra.hpsc.csiro.au/userguides/ now has a link in the "Getting Started" section to the University of Edinburgh UNIXhelp System.

Clicking on this takes you to a very useful beginners' guide to using UNIX (with some reservations - there are some features described which are University of Edinburgh only, and the man pages served will depend on the machine which the webserver runs on).

The UNIXhelp system has also been installed at the cherax.hpsc.csiro.au website, and this allows (searchable) access from your browser to the cherax man pages. Go to the Altix User Guide at http://intra.hpsc.csiro.au/userguides/ax/ and follow the link to "Other Documentation" near the bottom of the Table of Contents on the left. Clicking the link "Man pages are available" will give you access to the cherax man pages.

Phil Tannenbaum will present "Things You Never Wanted to Know about the SX-6, But Which Are Important Anyway" in the BMRC 9E meeting room on Thursday 18 August, 11:00-12:00. The talk will cover selected aspects of SX-6 memory operation, how I/O really works behind that READ and WRITE statement, and aspects of system scheduling that users should be aware of as more multi-node jobs become common.